[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20230811150947.18528aca@kernel.org>
Date: Fri, 11 Aug 2023 15:09:47 -0700
From: Jakub Kicinski <kuba@...nel.org>
To: Thinh Tran <thinhtr@...ux.vnet.ibm.com>
Cc: aelior@...vell.com, davem@...emloft.net, edumazet@...gle.com,
manishc@...vell.com, netdev@...r.kernel.org, pabeni@...hat.com,
skalluru@...vell.com, VENKATA.SAI.DUGGI@....com, Abdul Haleem
<abdhalee@...ibm.com>, David Christensen <drc@...ux.vnet.ibm.com>, Simon
Horman <simon.horman@...igine.com>
Subject: Re: [Patch v5 0/4] bnx2x: Fix error recovering in switch
configuration
On Fri, 11 Aug 2023 15:15:08 -0500 Thinh Tran wrote:
> As the BCM57810 and other I/O adapters are connected
> through a PCIe switch, the bnx2x driver causes unexpected
> system hang/crash while handling PCIe switch errors, if
> its error handler is called after other drivers' handlers.
>
> In this case, after numbers of bnx2x_tx_timout(), the
> bnx2x_nic_unload() is called, frees up resources and
> calls bnx2x_napi_disable(). Then when EEH calls its
> error handler, the bnx2x_io_error_detected() and
> bnx2x_io_slot_reset() also calling bnx2x_napi_disable()
> and freeing the resources.
It's looking fairly reasonable, thanks!
We do need commit messages describing motivations and impact
of each individual commit, tho, and those commits which are
not refactoring / improvements but prevent crashes need to
have a Fixes tag pointing to the first commit in history where
problem may occur (sometimes it's the first commit of the entire
history).
--
pw-bot: cr
Powered by blists - more mailing lists